Cloud side-end collaborative IVCPS multi-user concurrent traffic task joint simulation method

Abstract

The invention belongs to the technical field of intelligent automobiles, and discloses a cloud-side-end cooperative IVCPS multi-user concurrent traffic task joint simulation method, which comprises a terminal layer, an edge layer, a cloud layer and a performance monitoring module, and is characterized in that a cloud-side-end cooperative IVCPS multi-user concurrent simulation architecture is established by using a plurality of user terminals and a server, and the cloud-side-end cooperative IVCPS multi-user concurrent traffic task joint simulation is realized. Different users can individually edit traffic simulation tasks on the terminal layer, the tasks are submitted to the edge layer, the tasks are classified and arranged by the edge layer and then uploaded to an algorithm processing module of the edge cloud or the cloud end, and execution results are returned to the terminal layer. Meanwhile, in the simulation process, the performance index of the cloud side end is monitored in real time. According to the simulation and simulation based on the traffic concurrence scene, IVCPS multi-user concurrence access cloud side end service under cloud side end cooperation can be completed, and research on a coupling mechanism and development verification and evaluation of an automatic driving algorithm are facilitated.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent vehicles, and specifically relates to a cloud-edge-end collaborative IVCPS multi-user concurrent traffic task joint simulation method. Background Art

[0002] In the IVCPS system, concurrent tasks are mainly initiated by multiple intelligent vehicles according to their own needs, which simultaneously initiate requests to the cloud layer and edge cloud. After receiving and processing the relevant requests, the cloud layer and edge layer return the decision to the terminal layer. Specifically, intelligent connected vehicles will encounter various working conditions during driving. Under the cloud-edge-end collaborative architecture, considering the limited resource scale of the vehicle side, the control decision module is arranged in the cloud layer and edge layer. The vehicle side uploads perception information, accepts and executes corresponding decisions, reduces the resource load of multi-user terminals, and the cloud-edge-end collaboration is responsible for real-time performance.

[0003] However, there is currently a lack of relevant research on the coupling mechanism between concurrent tasks under IVCPS. Therefore, the present invention proposes a joint simulation method for multi-user concurrent traffic tasks of IVCPS in cloud-edge-end collaboration to simulate and verify the concurrent traffic scenarios of IVCPS under cloud-edge-end collaboration. Summary of the invention

[0004] In view of this, the present invention provides a joint simulation method for multi-user concurrent traffic tasks of IVCPS in cloud-edge-end collaboration. The present invention aims to further study the coupling mechanism between concurrent tasks under IVCPS, and simulate and verify the concurrent traffic scenarios of IVCPS under cloud-edge-end collaboration.

[0005] To achieve the above object, the present invention provides a cloud-edge-device collaborative IVCPS multi-user concurrent traffic task joint simulation method, comprising the following steps:

[0006] S1. Use multiple user terminals and servers to build the IVCPS simulation system architecture;

[0007] The IVCPS simulation system architecture includes a terminal layer, an edge layer, a cloud layer, and a performance monitoring module;

[0008] S2. Multiple terminal users who conduct online simulation or simulate real traffic scenarios through the terminal layer, that is, customized configuration is performed at the terminal layer according to the traffic simulation task;

[0009] S3. The traffic simulation tasks uploaded by the terminal layer are arranged and distributed through the edge layer, and the concurrent task subsets with low latency requirements and concurrent task sets without low latency requirements are obtained;

[0010] The edge layer is provided with a low-latency requirement autonomous driving algorithm module, to which a low-latency requirement concurrent task subset is sent, and then a decision result is formed and returned to the corresponding multi-user terminal;

[0011] The cloud layer is provided with an automatic driving algorithm module with non-low latency requirements, and a concurrent task set with non-low latency requirements is sent to the automatic driving algorithm module with non-low latency requirements, and then a decision result is formed and returned to the corresponding multi-user terminal;

[0012] S4. The performance monitoring module monitors the CPU, memory, throughput and I / O utilization of multi-user terminals, edge layer servers and cloud layer servers in real time, and monitors the communication performance status between the terminal layer, edge layer and cloud layer.

[0013] Furthermore, in step S1, the terminal layer includes the following contents:

[0014] Road topology establishment: used for users to create road networks according to their needs, including road type, number of lanes and road curvature;

[0015] Vehicle model customization: used for users to add vehicle models and set parameters of vehicle model, size, and speed as needed;

[0016] Smart sensor model: used for users to add radar and camera sensor models to simulate vehicle perception of the environment;

[0017] Driver Model: allows users to add different types of driver models to simulate different driving behaviors and decisions;

[0018] Vehicle dynamics model: The system provides a vehicle dynamics model for users to adjust the acceleration and steering parameters of the vehicle to simulate real vehicle driving behavior.

[0019] Furthermore, Simulink is deployed in the edge layer. During the simulation, the network communication module UDP Send or TCP / IP Send module in Simulink is used to process and package the perception information, and the packaged perception information is uploaded to the edge layer.

[0020] Furthermore, the concurrent tasks with low latency requirements and the concurrent tasks with non-low latency requirements in step S3 are defined by setting the task requirement response time as a threshold. When the task requirement response time is greater than the threshold, it is defined as a concurrent task with non-low latency requirements, otherwise it is a concurrent task with low latency requirements.

[0021] Further, the low latency requirement autonomous driving algorithm module in step S3 includes an emergency braking warning algorithm, a speed limit warning algorithm, a forward collision warning algorithm, a left turn assist algorithm, and a blind spot warning / lane change warning algorithm;

[0022] Emergency Braking Warning Algorithm: It uses vehicle sensors to monitor the environment and other vehicles around the vehicle. When a potential collision risk is detected, the system will warn the driver to take emergency braking or evasive measures;

[0023] Speed ​​limit warning algorithm: It uses vehicle positioning and map data, combined with the speed sensor, to identify the speed limit information of the road where the vehicle is currently located. Once the vehicle exceeds the speed limit, the system will issue a warning to the driver, reminding the driver to slow down to comply with the road speed limit;

[0024] Forward Collision Warning Algorithm: It uses the vehicle's sensors to monitor the road conditions and distance ahead, and predicts the possible risk of collision based on the vehicle speed and the speed of the vehicle ahead. When a potential collision threat is detected, the system will warn the driver, prompting the driver to take action to avoid the collision;

[0025] Left-turn assist algorithm: Usually used in conjunction with the blind spot warning algorithm, it monitors the blind spots around the vehicle through sensors on the side of the vehicle and provides a warning when the driver intends to turn left to avoid collisions with vehicles traveling sideways;

[0026] Blind spot warning / lane change warning algorithm: uses sensors on the side of the vehicle to monitor blind spots. The system will issue a warning to the driver, reminding the driver to pay attention to whether there are other vehicles on the side to ensure the safety of lane change operations.

[0027] Furthermore, the non-low latency requirement autonomous driving algorithm module in step S3 includes a path planning algorithm and a congestion reminder algorithm;

[0028] Path planning algorithm: It considers road distance and traffic conditions to determine the best driving path for the vehicle from the starting point to the destination;

[0029] Congestion Alert Algorithm: Based on real-time traffic data and historical traffic patterns, it detects road congestion and sends alerts to drivers, suggesting alternative routes or delaying the trip.

[0030] Furthermore, in step S4, the performance monitoring module uses wireshark as a network packet capture tool to capture the communication between the terminal layer, edge layer and cloud layer in real time, and uses the capture filter to capture and analyze the communication of UPD or TCP / IP communication protocol to analyze the communication performance between the terminal layer, edge layer and cloud layer.

[0031] Furthermore, the analysis in the performance monitoring module includes delay analysis, packet loss rate analysis, bandwidth occupancy rate analysis, traffic pattern analysis and connection establishment time analysis.

[0032] Beneficial effects:

[0033] The present invention discloses a method for jointly simulating multi-user concurrent traffic tasks of IVCPS in collaboration with cloud-edge terminals, including a terminal layer, an edge layer, a cloud layer and a performance monitoring module. A multi-user concurrent simulation architecture of IVCPS in collaboration with cloud-edge terminals is built by using multiple user terminals and servers. Different users can edit traffic simulation tasks in a personalized way at the terminal layer. Tasks are submitted to the edge layer, which classifies and arranges them and uploads them to the algorithm processing module of the edge cloud or cloud. The execution results are returned to the terminal layer. At the same time, during the simulation process, the performance indicators of the cloud-edge terminal are monitored in real time. The above simulation and emulation based on concurrent traffic scenarios can complete the concurrent access of multi-user IVCPS to cloud-edge terminal services under the collaboration of cloud-edge terminals, which is conducive to the study of its coupling mechanism and the development, verification and evaluation of autonomous driving algorithms.

[0034] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is an architecture diagram of a cloud-edge-end collaborative IVCPS multi-user concurrent traffic task joint simulation method of the present invention;

[0036] Figure 2 It is a multi-user terminal layer information interaction diagram;

[0037] Figure 3 This is the architecture diagram for concurrent task processing at the edge layer;

[0038] Figure 4 This is a flow chart of Example 1. DETAILED DESCRIPTION

[0039] In order to make the technical solutions, advantages and purposes of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of this application.

[0040] like Figure 1 As shown, the present invention provides a cloud-edge-end collaborative IVCPS multi-user concurrent traffic task joint simulation method, comprising the following steps:

[0041] S1. Use multiple user terminals and servers to build the IVCPS simulation system architecture;

[0042] The IVCPS simulation system architecture includes the terminal layer, edge layer, cloud layer, and performance monitoring module;

[0043] S2. Multiple terminal users who conduct online simulation or simulate real traffic scenarios through the terminal layer, that is, customized configuration is performed at the terminal layer according to the traffic simulation task;

[0044] like Figure 2 As shown, based on the Prescan design software, users create and adjust traffic scenarios, establish road topology structures according to requirements, and add vehicle models, intelligent sensor models, driver models, and vehicle dynamics models to meet the needs of different algorithms for vehicle perception and decision execution. The details are as follows:

[0045] 1) Road topology establishment: Users can create road networks according to their needs, including road type, number of lanes, road curvature, etc.;

[0046] 2) Vehicle model customization: Users can add vehicle models and set vehicle parameters such as vehicle type, size, speed, etc. as needed;

[0047] 3) Intelligent sensor model: Users can add various sensor models, such as radar, camera, etc., to simulate the vehicle's perception of the environment;

[0048] 4) Driver Model: Users can add different types of driver models to simulate different driving behaviors and decisions.

[0049] 5) Vehicle dynamics model: The system provides a vehicle dynamics model, and users can adjust vehicle acceleration, steering and other parameters to simulate real vehicle driving behavior;

[0050] During the simulation, multiple users simultaneously process and package the perception information in Simulink to ensure that the data format is correct and complete. They use the network communication modules in Simulink, such as UDP Send or TCP / IP Send modules, to upload the packaged perception information to the edge layer server, forming a concurrent task set {t1, t2, t3, t4, t5, t6, t7, ..., tm};

[0051] The decision results of the edge cloud and cloud-side algorithms are used as input and connected to the vehicle driver and dynamics model for execution.

[0052] S3. The traffic simulation tasks uploaded by the terminal layer are arranged and distributed through the edge layer, and the concurrent task subsets with low latency requirements and concurrent task sets without low latency requirements are obtained;

[0053] The edge layer is equipped with a scheduling processing module, which is mainly used for scheduling and allocating tasks;

[0054] The concurrent tasks with low latency requirements and the concurrent tasks with non-low latency requirements in step S3 are defined by setting the task requirement response time as a threshold. When the task requirement response time is greater than the threshold, it is defined as a concurrent task with non-low latency requirements and is sent to the cloud for calculation and processing through the scheduling processing module of the edge layer; otherwise, it is a concurrent task with low latency requirements and is directly handed over to the algorithm module deployed in the edge layer for calculation and processing;

[0055] The edge layer is equipped with a low-latency autonomous driving algorithm module, to which a low-latency concurrent task subset is sent, and then a decision result is formed and returned to the corresponding multi-user terminal;

[0056] like Figure 3 The edge layer concurrent task processing architecture shown in the figure. The Simulink environment is deployed in the edge server to establish an edge layer structure, which is mainly used to receive tasks uploaded from the terminal and classify and distribute these tasks based on the corresponding algorithm.

[0057] This task set can be divided into two subsets: the concurrent task subset with low latency requirements {t1, t2, t3, ..., tn} and the concurrent task subset with non-low latency requirements {tn+1, tn+2, tn+3, ..., tn+m}. The concurrent task set with non-low latency requirements will use the network communication modules in Simulink, such as UDP Send or TCP / IP Send modules, to send tasks to the cloud server, while the concurrent task subset with low latency requirements will be sent to the low latency requirement autonomous driving algorithm module.

[0058] Low latency requires autonomous driving algorithm modules. It covers multiple key functional modules, including emergency braking warning algorithm, speed limit warning algorithm, forward collision warning algorithm, left turn assist algorithm, blind spot warning / lane change warning algorithm, etc.

[0059] 1) Emergency braking warning algorithm uses vehicle sensors (such as radar and cameras) to monitor the environment and other vehicles around the vehicle. When a potential collision risk is detected, the system will warn the driver to take emergency braking or avoidance measures;

[0060] 2) Speed ​​limit warning algorithm uses vehicle positioning and map data, combined with vehicle speed sensors, to identify the speed limit information of the road where the vehicle is currently located. Once the vehicle exceeds the speed limit, the system will issue a warning to the driver to remind him to slow down to comply with the road speed limit;

[0061] 3) Forward collision warning algorithm uses the vehicle's sensors to monitor the road conditions and distance ahead, and predicts the possible risk of collision based on the vehicle speed and the speed of the vehicle ahead. When a potential collision threat is detected, the system will warn the driver to take action to avoid the collision;

[0062] 4) Left turn assist algorithms are usually used in conjunction with blind spot detection systems, which monitor the blind spots around the vehicle through sensors on the side of the vehicle and provide warnings when the driver intends to turn left to avoid collisions with vehicles traveling sideways;

[0063] 5) Blind spot warning / lane change warning algorithm uses sensors on the side of the vehicle to monitor blind spots, especially when the vehicle changes lanes. The system will warn the driver to pay attention to whether there are other vehicles on the side to ensure the safety of lane change operations;

[0064] After the low-latency concurrent task information data packet is decompressed, the vehicle's own information including speed, position, steering angle, etc., as well as the traffic environment information perceived by the vehicle sensor will be connected to the algorithm module as input. After calculation, the brake, throttle, and steering control information are output and packaged to form the decision result {d1, d2, d3, ..., dn}. The network communication module UDP Send or TCP / IP Send module in Simulink is used to return to the corresponding multi-user terminal, and after unpacking, it is used as input and connected to the vehicle driver and dynamics model for execution.

[0065] The cloud layer is equipped with an autonomous driving algorithm module with non-low latency requirements. The concurrent task set with non-low latency requirements is sent to the autonomous driving algorithm module with non-low latency requirements, and then a decision result is formed and returned to the corresponding multi-user terminal.

[0066] The cloud's non-low latency requires an autonomous driving algorithm module, which covers multiple key functional modules, including path planning algorithms, congestion reminder algorithms, etc.

[0067] 1) Path planning algorithm, which considers various factors such as road distance and traffic conditions to determine the best driving path for the vehicle from the starting point to the destination;

[0068] 2) Congestion warning algorithm, based on real-time traffic data and historical traffic patterns, detects road congestion and sends reminders to drivers, suggesting alternative routes or delaying the trip.

[0069] Use the network communication module UDP Receive or TCP / IP Receive module in Simulink to receive the concurrent subset {tn+1, tn+2, tn+3, ..., tn+m} with non-low latency requirements sent from the edge server. After the information data packet is decompressed, the vehicle's own information including speed, position, steering angle, etc. and the traffic environment information perceived by the vehicle sensor will be connected to the algorithm module as input, and the brake, throttle, and steering control information will be output after calculation. After packaging, the decision result {dn+1, dn+2, dn+3, ..., dn+m} is formed and returned to the corresponding client terminal for execution.

[0070] S4. The performance monitoring module monitors the CPU, memory, throughput and I / O utilization of multi-user terminals, edge layer servers and cloud layer servers in real time, and monitors the communication performance status between the terminal layer, edge layer and cloud layer.

[0071] The system performance monitoring module mainly uses Wireshark as a network packet capture tool to capture the communication between the cloud, edge and end in real time. It uses the capture filter to capture and analyze the communication of UPD or TCP / IP communication protocol, and analyzes the communication performance between the cloud, edge and end, including:

[0072] Delay analysis: Calculate the round-trip time (RTT) of data packets, evaluate the real-time and responsiveness of communications, and analyze transmission delays to understand the transmission speed of data packets in the network;

[0073] Packet loss rate analysis: Analyze the loss of captured data packets, calculate the packet loss rate, evaluate the reliability and stability of communications, and identify communication problems caused by packet loss;

[0074] Bandwidth occupancy analysis: Analyze the traffic pattern and data volume of captured data packets, calculate bandwidth occupancy, and evaluate communication load and network resource utilization;

[0075] Traffic pattern analysis: Analyze the captured communication traffic patterns, including the source, destination, protocol type and other information of the traffic;

[0076] Connection establishment time analysis: Track the handshake data packets during the connection establishment process, analyze the connection establishment time, and evaluate the communication establishment efficiency;

[0077] Communication performance monitoring mainly involves installing the PerfMon Metrics Collector plug-in in Jmeter, adding a test plan, and selecting "Add"->"Listener"->"jp@gc-PerfMon MetricsCollector" in the thread group of the test plan. In the tree view on the left, expand the "jp@gc-PerfMon Metrics Collector" node, select "PerfMon Metrics Collector", and then add the IP addresses and port number information of multi-user client terminals, edge servers, and cloud servers in the properties window to monitor performance indicators such as CPU, memory, throughput, and I / O utilization in real time. After the simulation is completed, you can export a CSV file for further analysis.

[0078] Example 1

[0079] like Figure 4 As shown, it is taken as an example that user A develops the AEB algorithm and user B develops the path planning algorithm.

[0080] Users A and B install Prescan software on their respective computers, draw road networks based on actual road information, and edit relevant vehicle information. After the settings are completed, users package the data and upload it to the edge layer. According to the attributes of the tasks, the tasks are assigned to different processing layers for calculation.

[0081] User A: AEB simulation task processing flow

[0082] 1. Data upload to the edge layer: Since the AEB (automatic emergency braking) simulation task has high latency requirements, it is directly connected to the low-latency algorithm library of the edge layer.

[0083] 2. Algorithm calculation: AEB algorithm calculation is performed in the low-latency algorithm library of the edge layer.

[0084] 3. Result return: The algorithm calculation results will be quickly returned to user A’s computer.

[0085] 4. Demonstration results: User A demonstrates the simulation results through PrescanViewer and observes the vehicle's response and performance in emergency situations.

[0086] User B: Path planning algorithm processing flow

[0087] 1. Data is sent to the cloud server: Since the path planning algorithm has low latency requirements, the data will be sent to the cloud server via UDP or TCP / IP communication protocol.

[0088] 2. Access the non-low-latency algorithm library: Data is accessed to the non-low-latency algorithm library in the cloud server to perform path planning algorithm calculations.

[0089] 3. Result return: The algorithm calculation results will be sent back to user B’s computer via the network.

[0090] 4. Demonstration results: User B demonstrates the simulation results through PrescanViewer to check the effect and efficiency of vehicle path planning.

[0091] During this process, Wireshark is used to capture the communication between the cloud and the edge in real time. The specific steps are as follows:

[0092] 1. Start Wireshark: Start the Wireshark software on the terminal computer.

[0093] 2. Select Network Interface: Select an appropriate network interface to capture traffic, such as an Ethernet interface or a wireless network interface.

[0094] 3. Set capture filter: Enter the filter rules of the target IP address and communication protocol type in the capture filter field of Wireshark. For example:

[0095] If monitoring UDP communication, you can enter ip.addr==<destination IP address>&&udp.

[0096] If you monitor TCP communication, you can enter ip.addr==<target IP address>&&tcp.

[0097] By setting a capture filter, Wireshark will only capture and display UDP or TCP communication packets related to the target IP address.

[0098] 4. Start packet capture: Click the "Start" button on the Wireshark interface to start packet capture.

[0099] 5. Analyze the communication status: During the packet capture process, view the captured data packets in real time, focusing on parameters such as the sending and receiving status of the data packets, delay, and packet loss rate. You can use various analysis tools of Wireshark, such as flow graphs and statistics, to deeply analyze the communication status between the cloud and the edge.

[0100] 6. Save the packet capture results: After the packet capture is completed, the captured data can be saved as a pcap file for subsequent analysis and archiving.

[0101] Use JMeter to monitor the corresponding CPU, memory, throughput and other parameters through the terminal computer IP, edge layer server IP and cloud server IP. The specific steps are as follows:

[0102] 1. Add a test plan and select "Add" -> "Listener" -> "jp@gc-PerfMon Metrics Collector" in the thread group of the test plan.

[0103] 2. In the tree view on the left, expand the jp@gc-PerfMon Metrics Collector node and select PerfMon Metrics Collector.

[0104] 3. In the Properties window, add the A and B user terminal computer IPs, edge layer server IPs, and cloud server IPs and port number information.

[0105] 4. Real-time monitoring of performance indicators such as CPU, memory, throughput, and I / O utilization.

[0106] 5. After the simulation is finished, you can export a CSV file for further analysis.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the protection scope of the present invention.

Claims

1. A cloud-edge-device collaborative IVCPS multi-user concurrent traffic task joint simulation method, It is characterized in that The following steps are involved: S1. Use multiple user terminals and servers to build the IVCPS simulation system architecture; The IVCPS simulation system architecture includes a terminal layer, an edge layer, a cloud layer, and a performance monitoring module; S2. Multiple terminal users who conduct online simulation or simulate real traffic scenarios through the terminal layer, that is, customized configuration is performed at the terminal layer according to the traffic simulation task; S3. The traffic simulation tasks uploaded by the terminal layer are arranged and distributed through the edge layer, and the concurrent task subsets with low latency requirements and concurrent task sets without low latency requirements are obtained; The edge layer is provided with a low-latency requirement autonomous driving algorithm module, to which a low-latency requirement concurrent task subset is sent, and then a decision result is formed and returned to the corresponding multi-user terminal; The cloud layer is provided with an automatic driving algorithm module with non-low latency requirements, and a concurrent task set with non-low latency requirements is sent to the automatic driving algorithm module with non-low latency requirements, and then a decision result is formed and returned to the corresponding multi-user terminal; S4. The performance monitoring module monitors the CPU, memory, throughput and I / O utilization of multi-user terminals, edge layer servers and cloud layer servers in real time, and monitors the communication performance status between the terminal layer, edge layer and cloud layer.

2. According to the cloud-edge-device collaborative IVCPS multi-user concurrent traffic task joint simulation method of claim 1, Features: In step S1, the terminal layer includes the following contents: Road topology establishment: used for users to create road networks according to their needs, including road type, number of lanes and road curvature; Vehicle model customization: used for users to add vehicle models and set parameters of vehicle model, size, and speed as needed; Smart sensor model: used for users to add radar and camera sensor models to simulate vehicle perception of the environment; Driver Model: allows users to add different types of driver models to simulate different driving behaviors and decisions; Vehicle dynamics model: The system provides a vehicle dynamics model for users to adjust the acceleration and steering parameters of the vehicle to simulate real vehicle driving behavior.

3. According to claim 2, a cloud-edge-end collaborative IVCPS multi-user concurrent traffic task joint simulation method, Features: Simulink is deployed in the edge layer. During the simulation process, the network communication module UDP Send or TCP / IP Send module in Simulink is used to process and package the perception information, and the packaged perception information is uploaded to the edge layer.

4. According to claim 3, a cloud-edge-device collaborative IVCPS multi-user concurrent traffic task joint simulation method, Features: The concurrent tasks with low latency requirements and the concurrent tasks with non-low latency requirements in step S3 are defined by setting the task required response time as a threshold. When the task required response time is greater than the threshold, it is defined as a concurrent task with non-low latency requirements, otherwise it is a concurrent task with low latency requirements.

5. According to claim 4, a cloud-edge-device collaborative IVCPS multi-user concurrent traffic task joint simulation method, Features: The low latency requirement autonomous driving algorithm module in step S3 includes an emergency braking warning algorithm, a speed limit warning algorithm, a forward collision warning algorithm, a left turn assist algorithm, and a blind spot warning / lane change warning algorithm; Emergency Braking Warning Algorithm: It uses vehicle sensors to monitor the environment and other vehicles around the vehicle. When a potential collision risk is detected, the system will warn the driver to take emergency braking or evasive measures; Speed ​​limit warning algorithm: It uses vehicle positioning and map data, combined with the speed sensor, to identify the speed limit information of the road where the vehicle is currently located. Once the vehicle exceeds the speed limit, the system will issue a warning to the driver, reminding the driver to slow down to comply with the road speed limit; Forward Collision Warning Algorithm: It uses the vehicle's sensors to monitor the road conditions and distance ahead, and predicts the possible risk of collision based on the vehicle speed and the speed of the vehicle ahead. When a potential collision threat is detected, the system will warn the driver, prompting the driver to take action to avoid the collision; Left-turn assist algorithm: Usually used in conjunction with the blind spot warning algorithm, it monitors the blind spots around the vehicle through sensors on the side of the vehicle and provides a warning when the driver intends to turn left to avoid collisions with vehicles traveling sideways; Blind spot warning / lane change warning algorithm: uses sensors on the side of the vehicle to monitor blind spots. The system will issue a warning to the driver, reminding the driver to pay attention to whether there are other vehicles on the side to ensure the safety of lane change operations.

6. According to claim 5, a cloud-edge-device collaborative IVCPS multi-user concurrent traffic task joint simulation method, Features: The non-low latency requirement autonomous driving algorithm module in step S3 includes a path planning algorithm and a congestion reminder algorithm; Path planning algorithm: It considers road distance and traffic conditions to determine the best driving path for the vehicle from the starting point to the destination; Congestion Alert Algorithm: Based on real-time traffic data and historical traffic patterns, it detects road congestion and sends alerts to drivers, suggesting alternative routes or delaying the trip.

7. A cloud-edge-device collaborative IVCPS multi-user concurrent traffic task joint simulation method according to claim 6, Features: In step S4, the performance monitoring module uses wireshark as a network packet capture tool to capture the communication between the terminal layer, edge layer and cloud layer in real time, and uses the capture filter to capture and analyze the communication of UPD or TCP / IP communication protocol to analyze the communication performance between the terminal layer, edge layer and cloud layer.

8. According to claim 7, a cloud-edge-device collaborative IVCPS multi-user concurrent traffic task joint simulation method, Features: The analysis in the performance monitoring module includes delay analysis, packet loss rate analysis, bandwidth occupancy rate analysis, traffic pattern analysis and connection establishment time analysis.

Images